When enabled, GraceBlocks’ AI capabilities are available across the platform, most notably within the Workflow manager. This feature lets you create actions that call generative AI prompts, using relational data from your GraceBlocks Zone and the AI model of your choice to generate image or text outputs. These outputs can then be mapped into fields in subsequent workflow actions.
AI functionality is also integrated into other areas of GraceBlocks, including the Chrome Plugin and during the creation of new tabs within a Block.
Whenever AI is triggered within GraceBlocks, usage is measured based on the pricing method of the selected model—such as tokens, characters, or other units. The specific pricing method and cost for each available model are detailed in the table below. are consumed from the Zone’s funds balance. Funds are used to cover both AI as well as native messaging (text and/or email) transactions. Each Zone is seeded with $2 to help you get started with both AI and native messaging.
To explore our subscription plans, click here.
The table below describes the models available throughout the GraceBlocks Platform and the affiliated Pricing:
Model description and utilization options | Pricing |
Gemini 2.0 Flash
Optimized for fast, cost-efficient performance, Gemini 2.0 Flash is great for quick text generation, summarization, and lightweight content tasks where speed is key. This model powers AI features across GraceBlocks, including the Run AI Process Workflow Action, the GraceBlocks Chrome Extension, new tab creation, and the AI prompt/response feature within the RTF Editor Field.
|
Pricing is applied both to the content prompt being used to instruct the AI as well as the results that are generated. Input Price: Results Price: In the context of AI models, a token is a chunk of text—typically a few characters or a short word. For example, the word "sunshine" might count as one token, while a longer sentence like "The quick brown fox jumps over the lazy dog." could be around 9–10 tokens. To put it in perspective, 1 million tokens is roughly equivalent to 750,000 words—the length of To Kill a Mockingbird seven times over.
|
Gemini 2.0 Flash Light**
A lighter, even more cost-effective variant of Gemini 2.0 Flash, ideal for basic prompts, short text completions, and utility-focused tasks with minimal processing needs. This model is available when leveraging the Run AI Process Workflow Action.
|
Pricing is applied both to the content prompt being used to instruct the AI as well as the results that are generated. Pricing is applied both to the content prompt being used to instruct the AI as well as the results that are generated. Input Price: Results Price: In the context of AI models, a token is a chunk of text—typically a few characters or a short word. For example, the word "sunshine" might count as one token, while a longer sentence like "The quick brown fox jumps over the lazy dog." could be around 9–10 tokens. To put it in perspective, 1 million tokens is roughly equivalent to 750,000 words—the length of To Kill a Mockingbird seven times over. |
Gemini 2.5 Flash Preview
|
Pricing is applied both to the content prompt being used to instruct the AI as well as the results that are generated. Input Price: Results Price: In the context of AI models, a token is a chunk of text—typically a few characters or a short word. For example, the word "sunshine" might count as one token, while a longer sentence like "The quick brown fox jumps over the lazy dog." could be around 9–10 tokens. To put it in perspective, 1 million tokens is roughly equivalent to 750,000 words—the length of To Kill a Mockingbird seven times over. |
Gemini 2.5 Pro Preview
|
Pricing is applied both to the content prompt being used to instruct the AI as well as the results that are generated. Input Price: Results Price: In the context of AI models, a token is a chunk of text—typically a few characters or a short word. For example, the word "sunshine" might count as one token, while a longer sentence like "The quick brown fox jumps over the lazy dog." could be around 9–10 tokens. To put it in perspective, 1 million tokens is roughly equivalent to 750,000 words—the length of To Kill a Mockingbird seven times over. |
Google Text to Speech
Transforms written text into natural-sounding spoken audio with a variety of output options ranging from standard to studio quality —ideal for accessibility use cases, voice-based applications, and automated readouts of content. Offers both male and female voices. This model is available for use in GraceBlocks through the AI speech to text process workflow action.
|
Pricing is determined by the number of characters in the input text and the selected voice type used to generate the audio. The list below outlines available voice types, their associated pricing, and audio samples to showcase voice quality. As a reference point, 1 million characters is approximately the length of the novel Moby-Dick, offering a helpful benchmark for estimating usage. When using the GraceBlocks workflow engine to convert text to audio, you’ll select a specific voice variant. Each variant name includes the language, locale, voice type, and gender. Example:
Voice TypesStandard voices: News voices:
US $0.000032 per character ($32 per 1 million characters)
Example from one of the News voices: WaveNet voices:
US $0.000032 per character ($32 per 1 million characters)
Example from one of many WaveNet voices: Neural2 voices:
US $0.000032 per character ($32 per 1 million characters)
Example from one of many Neural2 voices: Polyglot (Preview) voices: (aka Multilingual voices)
US $0.000032 per character ($32 per 1 million characters) Example from one of the available Polygot voices: Chirp voices:
US $0.00006 per character ($60 per 1 million characters)
Example from one of many Chirp voices: Studio voices:
US $0.00032 per character ($320 per 1 million characters)
Example from one of the available Studio voices: |
Google Vision
Analyzes images to extract text from a designated single attachment field, detect objects, or identify features—useful for processing photo uploads, automating data extraction, and powering image-driven workflows. This model is available for use in GraceBlocks through the AI image detection workflow action.
|
Charges are incurred per file. For files with multiple pages, such as PDF files, each page is treated as an individual image. Each feature applied to an image is a billable unit. The list below shows the price and a brief description of expected behavior for each feature per 1000 units. Label Detection: US $.003 per unit ($3.00 per 1000 units) Text Detection: US $.003 per unit ($3.00 per 1000 units) Document Text Detection: US $.003 per unit ($3.00 per 1000 units) Safe Search (explicit content) Detection: US $.003 per unit ($3.00 per 1000 units) Facial Detection: US $.003 per unit ($3.00 per 1000 units) Landmark Detection: US $.003 per unit ($3.00 per 1000 units) Logo Detection: US $.003 per unit ($3.00 per 1000 units) Image Properties: US $.003 per unit ($3.00 per 1000 units) Crop Hints: US $.003 per unit ($3.00 per 1000 units) Web Detection: US $.007 per unit ($7.00 per 1000 units) Object Localization: US $.0045 per unit ($4.50 per 1000 units)
|
GraceBlocks AI usage is governed by the GraceBlocks AI Terms and Conditions.